基于眼睛特征跟踪的眼睛状态跟踪_曹倩霞.pdfVIP

. 2007. 15( 12) # 1794 # Computer Measurement Control : 1671- 4587( 2007) 12- 1794- 04 : T P391141 : A 1 2 1 曹倩霞 , 罗大庸 , 李 顺 ( 1. , 4 10076; 2. , 410075) : , , , , , PII/ 500 , 256M , Win ow s 2000 7 214 , 97. 6% : ; ; Eye States Tracking Based on Tracking Eye Features 1 2 1 Cao Qianxia , Luo Dayong , Li Shun ( 1. Transportation College, Changsha Universit y of Science Technology , Changsha 4 10076, China; 2. College of Informat ion Science an Engineering , Central South University, Changsha 4 10075, China) Abstract: T here are many applications of the robu st eye states tracking, such as facial expression analysis an river aw areness system s. In these researches, w e nee to o more than tracking the locations of the person. s eyes but obtain their etaile escription. We nee to re- cover the state of the eyes an the parameters of an eye m o el. , w e evelop a ey e states tracking m o el base on tracking eye features by ex- tracting eye features w hich characterizing the eye states firstly, then tracking these features to track the eye states. Pr ocessing spee on a Pentium PII/ 500 PC is approximately 7 frames/ secon . In ex perim ental tests on 214 image sequences from subject, accurate tracking results are obtaine in 97. 6% of im age sequences. Key words: eye features etection; eye features tracking; eye states tracking. 0 , , , , ; , , ;